While many sudden cardiac arrests (SCA), are caused by ventricular fibrillation (VF) lasting several mins (long duration VF [LDVF]) before defibrillation, most studies of VF and defibrillation have been performed during the first min of VF (short duration VF [SDVF]) with internal defibrillation electrodes. LDVF differs markedly from SDVF and the distribution of the shock potential gradient for external defibrillation differs markedly from internal defibrillation. Spontaneous refibrillation following defibrillation is common after LDVF but not SDVF. The goal of this proposal is to use electrical mapping in dogs to obtain basic information about the mechanisms of LDVF and how they differ from SDVF, defibrillation of LDVF, and spontaneous refibrillation following LDVF, with the hope that this knowledge will lead to improved therapy for SCA. The specific aims will all be accomplished in the same experiments. Specific Aim 1. Investigate the mechanisms of maintenance of LDVF and compare them to those of SDVF maintenance. Hypotheses to be tested include (1) LDVF but not SDVF is maintained by wavefronts arising from Purkinje fibers, (2) LDVF is type I VF in which action potential duration restitution is present rather than type II VF in which conduction velocity restitution is important, and (3) conduction block in the working myocardium during VF is caused by refractoriness, low excitatory capability of the advancing wavefronts, regions of necrosis, or heterogeneous Connexin 43 expression. Specific Aim 2. Investigate the mechanisms of failed defibrillation following LDVF. Hypotheses will be tested that defibrillation following LDVF differs from that following SDVF in that (1) a higher minimum shock potential gradient is needed to defibrillate, (2) the myocardium is more sensitive to the damaging effects of shocks, or (3) the shock successfully halts LDVF but VF quickly recurs. The hypothesis will be tested that Purkinje activation occurs during the postshock pause before activation is recorded in working myocardium. Specific Aim 3. Investigate the mechanisms of spontaneous refibrillation occurring about 1 min following defibrillation for LDVF. Hypotheses will be tested that refibrillation arises from (1) a focus, (2) reentry, (3) Purkinje fibers, (4) papillary muscle insertions, (5) necrotic regions caused by ischemia during LDVF or (6) reperfusion caused by chest compressions during cardiopulmonary resuscitation.